At the 2022 World ADC Conference in San Diego, Stewart Mitchell, Executive Vice President and site head of Sterling Pharma Solutions’ Deeside, U.K. facility, presented a case study on the development and GMP manufacture at the site.
Sterling Pharma Solutions, a global Contract Development and Manufacturing Organisation (CDMO) with sites in the U.K. and the U.S., and established in small molecule development, identified a growth opportunity through the acquisition of a high potency biologics contract manufacturing facility, formerly known as ADC Bio in April 2021. Sterling has expanded rapidly in recent years, and with this purchase, took the chance to bring its funding, bandwidth and expertise to enable the growth of its new ADC business.
In Deeside, a 6,500 square-meter bespoke facility had recently been constructed, which included laboratories designed specifically to develop and perform highly potent bioconjugations. The experienced team of bioconjugation experts is capable of generating proof-of-concept materials at low milligram scale, in vivo-grade development material at milligram to gram scale, manufacturing toxicology batches, as well as process and analytical development. Until recently, this was all being undertaken for pre-clinical non-GMP projects, but post acquisition, Sterling has invested, recruited, and commissioned new equipment to permit the capability of manufacturing clinical GMP material at the site.
Proof of Concept
To ensure that systems and procedures were tested effectively, Sterling funded and executed an exemplar project with two aims: first, to demonstrate technical and GMP confidence to clients; and second, to demonstrate competence to the licensing authorities and to generate data for a Manufacturing and Import Authorisation (MIA(IMP)) license application. The company is currently awaiting audit and approval by the U.K. Medicines and Healthcare products Regulatory Agency (MHRA), which is expected in the first quarter of 2023.
Over a period of 12 months, which is typical for a GMP full development and ADC manufacture program, Sterling ran an “exemplar” project to develop and manufacture three GMP ADC drug product intermediate batches within a GMP program that would be suitable for Phase 1 clinical studies. The objective was to establish and develop a conjugation process and analytical methods to support the clinical manufacture of an ADC drug substance. Specifically, the project aimed to develop a robust and scalable process for the manufacture of an ADC consisting of a CD25-targeted antibody HuMax-TAC, conjugated to vedotin at an average drug to antibody ratio (DAR) of 4.
Figure 1.0: Exemplar project plan
The high-level project plan (figure 1.0) shows the timing for development of the process reactive stages, linking them with the purification stage during laboratory-scale demonstration runs and final execution of toxicology lots; after which the process was fixed and preparations for full-scale manufacture under GMP began. Ongoing throughout this was the development of analytical methods required for process and purification development, in-process controls, and product release testing. Material from the toxicology lots was used to qualify the analytical methods and establish a working reference standard prior to manufacture of the GMP batches.
In parallel, preparations were made for GMP manufacture, which was initiated in 2022 and resulted in the successful manufacture of three GMP bulk drug substance (BDS) batches, which were subsequently analyzed and released by the site Qualified Person (QP). The whole program of work including development data, analytical development, and batch manufacturing records. In-process and release testing records are now held on site for open book review to auditors and prospective clients to demonstrate competence and capability in GMP manufacturing.
Process Development
The process development phase of the project was initiated via a theoretical ‘paper’ process design based on the target specification and using the team’s knowledge and experience, of the antibody toxin and reaction chemistries. The first laboratory trials involved testing the reactive stages of the designed process and evaluation of established platform analytical methods for suitability for this specific ADC. Once process development and optimization were complete, the reactive stages were fixed, and the overall scalability assessed. At this stage, the team started to incorporate and evaluate tangential flow filtration (TFF) for removal of residual toxin and other process related contaminants by diafiltration, and clarification and bioburden reducing filtration steps within the process.
In parallel, the control strategy for each of the process parameters was devised and robustness studies performed, together with any support studies needed to aid transfer of methods and processes to quality control (QC) and manufacturing. When platform methods were proven to be unsuitable, further analytical optimization was carried out. Once development was complete, the process was fixed prior to execution of the toxicology and GMP batches.
The process had five main stages: a reduction reaction; a conjugation reaction, diafiltration purification, followed by formulation, and a final filtration and fill to give the final bulk drug product intermediate.
Initially, the project’s focus was on the reactive stages and employed a scaled down model with gel permeation purification used instead of TFF to evaluate these reactions at small scale. The key outputs were the percent monomer, the average DAR as determined by hydrophobic chromatography (HIC), and the percentage of DAR 1 and DAR 3 species in the HIC profile, which is an indicator of the efficiency of the conjugation reaction.
The trials of the designed process generated conjugate with quality attributes at or above target specification. The target DAR was within specification with very low levels of DAR 1 and 3 species, and the percentage of monomer was well above the 90% lower limit. The platform size exclusion chromatography (SEC) and HIC methods were also shown to be suitable for measurement of monomer, average DAR and extent of conjugation.
Process Optimization
The approach then became one of process optimization, where the focus was on gaining a better understanding of reaction stoichiometries and kinetics, starting with the reduction reaction. In this type of conjugation process, TCEP equivalents are the most critical process parameters controlling DAR, so their effect on the process needed to be understood.
The exact TCEP equivalents to achieve the target DAR were determined, as well as the TCEP equivalents at which the upper and lower ranges of the DAR specification were achieved, which was helpful in determining the control strategy for this parameter.
The reaction time was also studied, firstly to ensure that the chosen reaction time was on a kinetic plateau and there was a stable window of operation either side for manufacturing, and secondly, that the reduced intermediate was stable with no significant change in average DAR.
A similar set of experiments was performed on the conjugation reaction with the aim of determining whether the equivalents used during the conjugation reaction could be reduced without affecting average DAR or the percentage output of DAR 1 species. This would reduce not only raw materials but also the amount of residual free toxin that needed to be removed during the purification stage. A study of the conjugation kinetics showed that the chosen conjugation time of 5 minutes was on a kinetic plateau, with a 35-minute window of stability either side. (figure 2.0)
Figure 2.0: Process optimisation: conjugation time.
Once the conditions for the reduction and conjugation stages were fixed, a number of small-scale confirmation runs of fixed conditions were then performed by multiple operators with individual reagent preparations to confirm the optimized process was performing as expected, and to establish some basic robustness.
The next stage was to assess the scalability of the reaction conditions. A gram-scale model was used and the diafiltration purification and filtration steps added. As with the reactive stages, experience with clearance of residual vedotin by diafiltration informed selection of the TFF membrane and operating parameters such as membrane load, crossflow and transmembrane pressure. The key outputs for this study were the removal of residual toxin and solvent during the diafiltration stage; evaluation of the performance of the filtration stages; evaluation of process scalability in terms of average DAR and percentage of monomer; and to establish whether product quality was maintained through the diafiltration and filtration steps. The data showed that the selected diafiltration parameters were able to achieve the required level of free drug and solvent removal. (figure 3.0)
Figure 3.0: Product scalability: residual toxin and solvent removal
Two filtration stages were added to the process – one post diafiltration, and the other after final formulation. The filtration stages were screened in real time, and filter selection based on performance – throughput versus pressure – and biased towards operator safety. The resulting data were then used to size suitable filters for the toxicology and GMP batches.
The process was found to be both scalable and reproducible with consistent data achieved on duplicate runs. The conjugate generated from scale-up run 1 was transferred to QC and used to generate an interim reference standard for method validation.
In collaboration with the QC and manufacturing teams, the control strategy for each process parameter was established. A design of experiments (DOE) robustness study was performed on a number of key process parameters to show that the proposed control ranges were suitable. A fractional 2-level factorial design was chosen with variable ranges set at least two times the proposed control ranges, creating a design space around the process. The key outputs for this process were average DAR, percentage of DAR 1 species and percentage of monomer present.
Some variation in average DAR and percentage of DAR 1 species was observed across the design space, but in all reactions the average DAR remained within the specification range. The percentage of monomer also remained well above the lower specification limit, with greater than 99% achieved in all reactions.
This control strategy was used during the execution of two toxicology lots, which showed very similar product quality to the gram-scale runs. The performance of the diafiltration and filtration stages was further verified, with residuals clearance and process intermediate and product filterability scaling as predicted. Process solution characterization and shelf-life studies along with process intermediate hold time data were generated and a detailed process report and description provided to manufacturing to transfer the process to them.
Analytical Development
After a successful transfer to the QC labs, additional methods were introduced to cover all the relevant aspects of biologics, such as quality, safety, strength, potency and the major impurities related to the process. Some of the methods were pharmacopoeial, while others were transferred from process development or developed in QC.
There are several possible starting points for establishing analytical methods for an ADC. One is a transfer of a client method, which was not relevant in this case. The second approach is to use a platform method established in the company, then adapt and optimize it specifically for the candidate ADC, and this was the route used for the HIC methods. Another option is to develop a suitable method de novo, starting with method feasibility and development throughout optimization, and it was this route that was employed for the cell killing assay.
The same validation strategy was used for the two different method developments, but in addition to the phase-appropriate validation criteria tested, specificity, stability and robustness checks were introduced to make sure that the methods were not only suitable for release testing, but could also be used for stability testing.
Even when using a platform method, its suitability must be verified, and often minor adjustments are necessary to make the method meet all of the validation criteria. The HIC method was optimized by adding a cleaning cycle to stop any carryover of high DAR species, and by varying the protein load at various different detection wavelengths to determine that 214nm was the optimum wavelength.
Given the variability in cell lines, it is not considered possible to have a platform method for the cell killing assay or potency assay for a new ADC, because different toxins will affect the required dilution range and incubation times. The QC team developed a cell bank of CD25 expressors to gain information on the doubling times. With the prior knowledge of mode of action of the toxin in question, a basic assay format was developed. A range-finding experiment was then run, taking into account cell density, ADC load, incubation time, dye time and cell resuscitation time.
After developing a working potency assay specifically designed for the ADC, the effect of the concentration of ADC with different average DAR values on the relative potency was investigated. The assay proved to be sensitive enough to indicate changes of average DAR, and could be used to make more informed decisions on its use together with other DAR-sensitive methods, such as HIC. The assay was also validated against additional characteristics, such as specificity, stability and robustness.
Scale-up Success
A review of the exemplar project data showed a high degree of consistency across 6 pivotal process batches, from the time the process was fixed at gram-scale, across two toxicology batches executed in development and three GMP batches executed in the manufacturing suite. The data covered 50-fold scale-up across several months, with the process having been executed in various laboratories by different teams, using similar but not identical equipment. The runs also used multiple lots and batches of critical raw materials, and involved inherent variability in process solution preparation.
To date, the company has performed a successful scale-up from laboratory through to GMP manufacturing, without impacting safety and without introducing any changes to the product. Analytical methods have all been verified, and the average QC release time was four weeks for all three GMP batches. Currently a stability study on the final GMP batch is being carried out.
Author:
Corresponding Authors: Steward Mitchell Email: Stewart Mitchell
Key terms: ADC, bioconjugation, ADC Development, ADC Manufacturing, CDMO, Process Development, GMO, GMP bulk substance, Antibody-drug Conjugate
Published In: ADC Review| Journal of Antibody-drug Conjugates
DOI: https://doi.org/10.14229/jadc.2023.01.12.003.
How to cite:
Mitchell, S. Opportunities for ADC Development and Manufacturing- J. ADC. January 12, 2023. DOI: 10.14229/jadc.2023.01.12.003
Last Editorial Review: December 10, 2022
Article History:
Original Manuscript Received January 8, 2022
Review results received January 9, 2022
Manuscript accepted for publication January 12, 2023
Featured image: Antibodies Structure Photo courtesy: © 2016 – 2022 ANIRUDH on Unsplash
